A common misconception about the meanings of words like teaching and learning is that learning and teaching occur when the teachers tell the students what they need to know, and then the students remember it.
Those of us who have been students should know better -- very, very few people ever learn this way. We forget what we memorized for that test. We parrot stuff back to teachers without necessarily understanding what it is we're saying or why we're saying it. (Those freaks among us who remember all this stuff -- a trifling percentage of the population -- do very well on game shows, but with startling frequency, don't do so hot at creative or analytical work.)
When do we learn? When we have to figure something out for ourselves, we learn, and remember it well. When we have to explain something to someone else, we often find we learn it pretty well. When we use knowledge and apply it to problems, we learn pretty well, then, too.
What's interesting -- but often unnoticed -- about all of these situations is that they involve confusion. You start off confused about something, but work it out on your own, until you reach understanding, and then you know it forever (or until your next head injury).
One mark of a sharp, well-trained mind is that it's comfortable during moments of confusion, and has learned to see them as okay. But a lot of us treat confusion as a bad thing -- something to be avoided. Students don't want to be confused, and teachers often (figuratively) wring their hands in despair when they realize their students are confused about something. But in a class where learning is happening, some confusion is inevitable -- it's the first step in the learning process. First, you think Thomas Kuhn makes no sense, but you plug on and try to make sense of him. If you keep at it, eventually you get it, and then you've learned something. (I'm not saying that all confusion is good. Just as there are good and bad types of fat, some types and causes of confusion are terrible for you. But a lot of the stuff that causes complaints is good fat, or good confusion, and unfairly indicted.)
The best ways to eliminate confusion in a class are often bad for you: The teacher can have you do things you already know how to do, or she can give you such clear step-by-step instructions for everything that you can surf the class on autopilot with -- as Hermione Granger in Harry Potter and the Order of the Phoenix puts it -- "no need to think." If the teacher instead asks students to figure something out, confusion is inevitable, at least until learning sets in.
Some people think that procedural confusion -- confusion about what to do or how to do it -- is among the bad-fat confusions. I used to think that, but now I'm not so sure. Lately, I've been doing little experiments to see how well students learn things when instructions are vague and fuzzy (some of my readers will have doubtless noticed). I give students the sorts of missions they'll get in a real workplace ("Hey, Bob. Write me a press release!") and the sorts of instructions one usually gets in those environments ("How? Don't ask me. I don't know. Look it up somewhere.") I worked for more than a decade in the "real world" outside of academia, in government, in industry, in newsrooms, and this sort of thing is remarkably common. The people who get promoted, who do the best, are the ones who can manage in a sea of vague instructions, who can do solid, quality work without hand-holding. Generally, they're people who have learned the hard way that they are able to figure things out, if they really want to do so. They come up with their own instructions, and being their own masters, become masters. Even speaking for myself, I know that the stuff I've had to figure out in this way -- the instructions I've had to give myself -- are the most useful sets of instructions I've ever had.
So for about a year now, a few times a quarter, I throw students a fairly vague, work-style prompt, in this sort of spirit: "Hey, the president wants mission statements and department philosophies, in memo format, by 7 a.m. tomorrow. Go write one for us. ... No, I don't know what he's talking about either. Figure it out. And make it good." And then I see how they do. I grade easier on these than I do on other papers (because not to do so would be evil), and try to make sure they know where things went wrong afterwords, but for a while, at least, they have to come up with their own plans, instructions, and standards. My goal here isn't emotional security, but learning, either during the process or right after it, when they can see where things went wrong.